Space Situational Awareness on Dr Yang Yang

Optical Tracking and Observatory Automation

yang.yang16@unsw.edu.au (Dr Yang Yang) — Tue, 01 Oct 2024 00:00:00 +0000

Problem

Ground-based optical tracking is the cheapest way to keep custody of objects in high orbits, and the hardest data to work with. Passes are short, gaps run to days, and the objects that matter most are often close to the detection limit. An orbit fitted to that data can look precise and still be wrong, because the reported covariance reflects the fit rather than the observation geometry that produced it.

Capability

My team develop the full path from photons to orbit solutions: image processing that pulls astrometry from faint and streaked detections, association and initial orbit determination for short arcs, and estimation that propagates realistic uncertainty rather than an optimistic fit residual.

Alongside this, the UNSW Observatory has been automated for remote and programmatic operation, so tracking can be tasked against a target list instead of scheduled by hand.

Satellite polarimetry extends the same sensor from tracking towards object characterisation — inferring attitude and surface properties, not just position. This is the ongoing PhD project of Asad Rizvi, and it is at an earlier stage than the tracking and automation work above: the observing approach is being developed on the UNSW telescope, with research outputs still to come.

The UNSW Observatory telescope on its equatorial mount, with control computer, guide scope and imaging camera fitted.

Demonstrated result

The UNSW Observatory now runs automated, remotely controlled satellite tracking. The automation and control work was delivered as an undergraduate thesis project by Tejas Margapuram (slides).
A modular three-tier control system now drives the observatory, designed, implemented and deployed by undergraduate thesis student Darcy Faulkes: an ASCOM Alpaca server abstracting heterogeneous hardware, a backend for visibility prediction, scheduling and orchestration, and a web frontend. Remote control was verified against third-party clients, and a complete satellite observation was carried out autonomously, with no human intervention.
An end-to-end optical space surveillance and tracking pipeline runs from raw frames to orbit solutions (overview slides).
An automated satellite tracking control project was funded by the NSW Space Research Network for 2025–2026.
Astrometric analysis of satellite streak images was developed end to end as a supervised thesis project, extending the pipeline’s usable detection range.

The observatory tasking interface: satellites predicted to be visible on a given night, with sky track and altitude profile for the selected object and one-click schedule capture. This is the top tier of a three-tier architecture built by undergraduate thesis student Darcy Faulkes — an ASCOM Alpaca server abstracting the observatory’s heterogeneous hardware, a backend handling visibility prediction, scheduling and orchestration, and this web frontend.

External use

The observatory is used for research and for space engineering education at UNSW, and it is the sensor behind the tracking pilot offered to partners.

Next partnership opportunity

A tracking campaign against objects you nominate, returning astrometry, orbit solutions and a residual and accuracy report — enough to judge whether a taskable optical sensor belongs in your architecture. See Partner with us.

Satellite Manoeuvre Detection and Pattern of Life

yang.yang16@unsw.edu.au (Dr Yang Yang) — Fri, 01 Nov 2024 00:00:00 +0000

Problem

A geostationary satellite changes its behaviour and the people responsible for the catalogue find out late, or not at all. The hard part is not seeing that the residuals moved — it is deciding whether that movement was a real manoeuvre, a mismodelled solar radiation pressure, or simply a bad track. Get that call wrong in one direction and you lose custody of the object; get it wrong in the other and you generate alarms nobody trusts.

Doing this at the scale of the GEO belt makes it worse. Analyst-led assessment does not scale to hundreds of objects observed nightly, and detectors trained on assumed dynamics degrade quietly as the population and operating practices change around them.

Capability

We build manoeuvre detection and pattern-of-life identification for geostationary and geosynchronous satellites, grounded in observational data rather than simulated behaviour. Three parts:

A labelled optical dataset. The Optical Pattern of Life Analysis and Library (OPAL) dataset gives detectors something real to train and be evaluated against.
Detection methods. Image-based and time-series transformer models for identifying manoeuvres, benchmarked against conventional residual-threshold approaches.
A pattern-of-life framework. Integrating the above so that individual detections become a behavioural characterisation of an object over time, which is what an operator actually acts on.

Related characterisation work — satellite polarimetry through the UNSW Observatory, the ongoing PhD project of Asad Rizvi — supplies independent evidence about attitude and surface state to corroborate a detection.

The work is carried out by Michael Ling, research assistant on the NSW Space Research Network project, together with several students on supervised thesis projects covering manoeuvre modelling, track association and object characterisation.

Demonstrated result

Competitively funded as a NSW Space Research Network Pilot Research Project for 2025–2026, Pattern of Life Identification for Geosynchronous Satellites Using Transformer-Based AI Foundation Models, in collaboration with Macquarie University and the University of Adelaide.
A dedicated research position stood up to deliver the GEO pattern-of-life characterisation work.
Method development presented externally, including transformers and deep learning for satellite manoeuvre detection at the UNSW AI Symposium (slides).
Complementary manoeuvre-modelling work on factor graph optimisation for orbit determination with Gaussian approximation of impulsive manoeuvres, developed as a supervised thesis project.

External use

The project is co-delivered with Macquarie University under NSW Space Research Network funding, and it feeds the NSW Space Research Network white paper on Space Domain Awareness.

Next partnership opportunity

A manoeuvre and pattern-of-life assessment on your GEO or LEO time series: a labelled event list over the period you care about, and detection performance benchmarked against your current baseline, with false-alarm behaviour stated openly. See Partner with us.

Debris Family Classification

yang.yang16@unsw.edu.au (Dr Yang Yang) — Wed, 10 Dec 2025 00:00:00 +0000

Problem

When a satellite breaks up, the fragments enter the catalogue as hundreds of unattributed objects. Reconnecting them to their parent event matters for attribution, for understanding how the breakup happened, and for predicting where the cloud goes next.

Machine-learning methods using proper elements have made progress on this, but they inherit a quiet failure mode: a model trained on an outdated representation of the debris environment degrades as the population evolves around it. There is also a specific technical trap — normalising quaternion-set features for a neural network destroys the orbital size information the classifier needs, and the loss is invisible until accuracy is measured.

Capability

A computational pipeline that generates synthetic fragmentation data from explosive breakup events using a Standard Breakup Model, propagates it under a high-fidelity dynamical model, and extracts proper elements across three representations — modified equinoctial (MEE), Poincaré (PNC) and quaternion (QTN) sets. Neural networks are then trained on combinations of those element sets to decide whether a pair of fragments shares a parent.

Extending beyond the modified-equinoctial space used by previous approaches widens the dynamical fingerprint available to the classifier. The pipeline also includes an augmented quaternion representation, QTNp, which explicitly restores the semi-latus rectum lost during feature normalisation.

Demonstrated result

ROC-AUC improved from 0.789 to 0.858 in synthetic Starlink-like LEO experiments, comparing the joint MEE + PNC + QTN feature set against the MEE-only baseline, with corresponding gains in accuracy and F1.
The QTNp augmentation lifted quaternion-set accuracy from 0.31 to 0.60, by restoring the orbital size information that standard normalisation discards — identifying and closing a failure mode that had not previously been characterised.
Published as a preprint, arXiv:2512.08495, led by undergraduate researcher Michael Ling, and presented at the Australian Space Research Conference 2025 (slides).

External use

The method addresses a recognised gap in space sustainability and space domain awareness: maintaining attribution of fragmentation debris as the circumterrestrial environment evolves. The preprint is openly available, and the finding about quaternion-set normalisation applies to any classifier built on that representation, not only to this pipeline.

Next partnership opportunity

Applying the classifier to a real breakup event of interest to you, or extending it from synthetic training data to your catalogue. This fits most naturally as a sponsored thesis or co-funded project — see Partner with us.

Thermospheric Density and Reentry Prediction

yang.yang16@unsw.edu.au (Dr Yang Yang) — Sun, 01 Jun 2025 00:00:00 +0000

Problem

Below about 600 km, atmospheric drag dominates the error budget, and drag is only as good as the density model behind it. JB2008, NRLMSIS 2.1 and DTM2020 disagree with each other and with reality, worst during geomagnetic storms. Reentry windows stay wide until late, and conjunction screening inherits the same uncertainty.

Capability

Reduced-order density modelling calibrated against real orbital behaviour rather than against another model: Starlink ephemerides assimilated through an Unscented Kalman Filter, comparing linear (POD with DMDc) and nonlinear (diffusion maps with m-PCE) reduction, each initialised from the three empirical baselines. The aim is a model fast enough for operational tracking and space traffic management.

Demonstrated result

Early stage — no published accuracy figure yet. The framework and assimilation pipeline are built and running as MPhil research by Ruoyan (Arthur) Zhao; the comparison against the baselines is in progress.

External use

Density error is the shared root cause behind wide reentry windows and weak conjunction screening, and it degrades orbit determination for the tracking and manoeuvre-detection work on this site.

Next partnership opportunity

Calibrating the model against your tracking or ephemeris data, or applying it to a specific reentry problem. Best suited to a co-funded or sponsored project while the method matures — see Partner with us.